https://eonh1u.github.io/tags/%E5%88%86%E5%B8%83%E5%BC%8F%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/Recent content in 分布式机器学习 on eonh1u BlogHugozh-cnWed, 30 Dec 2020 10:59:43 +0000https://eonh1u.github.io/posts/nccl-test-%E7%8E%AF%E5%A2%83/Wed, 30 Dec 2020 10:59:43 +0000https://eonh1u.github.io/posts/nccl-test-%E7%8E%AF%E5%A2%83/<p>在之前的服务器上发现安装的nccl，不能正确的运行nccl test测试，在尝试了多种方法之后，发现需要重新编译安装nccl</p> <p><a href="https://github.com/nvidia/nccl">https://github.com/nvidia/nccl</a></p> <p>按照官方github上说明进行编译安装</p> <pre tabindex="0"><code>apt install build-essential devscripts debhelper fakeroot -y make -j pkg.debian.build dpkg -i build/pkg/deb/*.deb </code></pre>